انجمن کامپیوتر ایران

براي حل مسئلهي خوشه بندي دو دسته روش کلی وجود دارد: روشهاي مسطح و روشهاي سلسله مراتبی . مزیت روش هاي سلسله مراتبی به روشهاي مسطح در مواردي است که تعداد خوشهها از اول مشخص نمیباشد و همینطور هنگامیکه علاوه بر یافتن رابطه بین نمونهها علاقه مند به یافتن رابطهي بین خوشهها نیز می باشیم . یکی از روشهاي مفید جهت بهبود خوشه بندي، حل یک مسئله ي خوشه بندي به روشهاي مختلف و ترکیب نتایج میباشد. با توجه به اینکه در الگوریتمهاي سلسله مراتبی وابستگی زیادي به پارامترهاي مسئله وجود دارد، ترکیب خوشه بندي هاي سلسله مراتبی مختلف میتواند این وابستگی را کاهش دهد. در این مقاله برمبناي مفاهیم سیستمهاي ایمنی مصنوعی، روشی براي ترکیب خوشه بندي هاي سلسله مراتبی ارائه میدهیم. به این منظور، ابتدا به روشهاي مختلف خوشه بنديهاي مختلفی بهدست آورده و سپس جمعیتی از خوشه بنديهاي تصادفی شکل داده و با تعریف معیار شباهتی سعی میکنیم جمعیت را به سمت خوشه بندياي ببریم که بیشترین شباهت را به کلیهي خوشه بنديهاي پایه داشته و بنابراین، اطلاعات ساختاري کلیهي خوشه بنديهاي پایه را حتی الامکان حفظ نماید . نتایج ارائه شده در مقاله، کارا بودن این روش و بهبود ایجاد شده در خوشه بندي سلسله مراتبی را نشان میدهد.

In WDM all-optical networks, physical impairments degrade optical signal to noise ratio (OSNR). Therefore, the quality of a lightpath must be checked during connection setup. Since during a lightpath setup, it may inject crosstalk on all previously established lightpaths with which it shares links or nodes, OSNR for those lightpaths should be checked as well as the OSNR of the tentative lightpath. If the number of hops for a lightpath increases, the amount of OSNR degradation increases due to crosstalk. Hence, longer paths are subject to higher QoT degradation due to crosstalk. In this paper, we enhance OSNR degradation for different lightpaths with different hop numbers in such a way that OSNR degradation for long-hop lightpaths becomes closer to short-hop lightpaths, i.e., a fair OSNR degradation. We provide a mechanism to estimate crosstalk risk at the routing process by which we can find the nodes with positive crosstalk risk. Then, rerouting procedure is used to reduce the crosstalk risks of these nodes by migrating some of lightpaths that pass through the nodes.

A collection of web pages which are about a common topic and are created by individuals or any kind of associations that have a common interest on that specific topic is called a web community. Since at present, the size of the web is over 3 billion pages and it is still growing very fast, identification of web communities has become an increasingly hard task. In this paper, a method based on asynchronous cellular learning automata (ACLA) for identification of web communities is proposed. In the proposed method first an asynchronous cellular learning automaton is used to determine the related pages and their relevance degree (the relationship structure of web pages). For determination of relationship structure of web pages information about hyperlinks and the users’ behaviour in visiting the web pages are used. Then, an algorithm similar to the HITS algorithm is applied on the obtained structure to identify the web communities. One of the advantages of the proposed method is that the web community obtained using this method is not dependent on a specific web graph structure. To evaluate the proposed approach, it is implemented and the results are compared with the results obtained for two existing methods, HITS and a complete bipartite graph based method. Experimental results show the superiority of the proposed method.

Polynomials are one of the most powerful functions that have been used in many fields of mathematics such as curve fitting and regression. Low order polynomials are desired for their smoothness1, good local approximation and interpolation. Being smooth, they can be used to locally approximate almost any derivable function. This means that when linear functions fail in approximation (e.g. where the first order Taylor expansion equals zero) polynomial functions can be used in local approximation, such that one can achieve better estimations at extremums. In this paper, application of polynomial kernel functions in locally linear neurofuzzy models is shown. Using polynomial kernels in local models, better local approximations in prediction of chaotic time series such as Mackey-Glass is achieved, and the capability of the neurofuzzy network is enhanced.

Recently, a technique has been proposed for image hiding, that is based on block texture similarity where, blocks of secret image are compared with blocks of a set of cover images and the cover image with the most similar blocks to those of the secret image is selected as the best candidate cover image to conceal the secret image. In this paper, we propose a new image hiding method in which, the secret image is initially coded using a fuzzy coding/decoding method. By applying the fuzzy coder, each block of the secret image is compressed to a smaller block. In this way, after compressing the secret image to a smaller one, we hide it in a cover image. Obviously hiding a smaller secret image causes less distortion in the stego-image (the image that has secret image or data) and therefore higher quality stego-image is obtained. Consequently, the proposed method provides higher embedding rate and enhanced security.

This paper describes the theory and implementation of dynamic Bayesian networks in the context of speaker identification. Dynamic Bayesian networks provide a succinct and expressive graphical language for factoring joint probability distributions, and we begin by presenting the structures that are appropriate for doing speaker identification in clean and noisy environments. This approach is notable because it expresses an identification system using only the concepts of random variables and conditional probabilities. We present illustrative experiments in both clean and noisy environments and our experiments show that this new approach is very promising in the field of speaker identification.

In this paper a two step method based on Gauss- Newton and factor graphs algorithm is proposed for localization to enhance accuracy of localization. The Gauss-Newton algorithm is accurate method for positioning. The most important challenge of this method is senility to initial point; this problem is solved in positioning based on factor graphs. So, in this paper, first positioning equations using angle of arrival is considered based on factor graphs algorithm. Second, final location estimation is performed using Gauss Newton algorithm with error near to Cramer-Rao bound. Simulation results shows that positioning error using two step method has maximum 6% gap to Cramer-Rao Bound.